Clustering and rough set-based knowledge discovery for product family planning

Product family planning has received much attention from both academia and industries. It aims at incorporating customers' needs into design elements of product family. The main challenger for product family planning originates from difficulties in mapping customer needs to product family specifications. This paper intends to develop a method to improve the mapping process by reusing knowledge from purchased products according to the satisfied customer needs. A knowledge discovery model for product family planning is proposed, where clustering is adopted to partition the purchased products so that commonality of product family could be effectively addressed and rough set is employed to extract the more concise decision rules. A case study of air condition is reported to illustrate the feasibility of proposed approach and associated algorithms.